JAGES10-13 cohort 0 Probabilistic inference on the health status of older people focusing on the conditional probability of the built environment and social capital-a prospective cohort study using the data of JAGES10- 13 cohort on the primordial prevention of community development / 建造環境および社会関係資本の組み合わせ条件に着目した高齢者の健康状態に関する確率推論 JAGES10-13 cohort を用いたまちづくりによる0 次予防の効果推定
- Resource Type
- Journal Article
- Authors
- Katsunori KONDO; Masamichi HANAZATO; Yoshitaka OTSUKA; 大塚 芳嵩; 花里 真道; 近藤 克則
- Source
- 環境情報科学 / Environmental Information Science. 2022, 51(2):98
- Subject
- Bayesian network
built environment
elderly people
pLSA
social capital
ベイジアンネットワーク
建造環境
社会関係資本
高齢者
- Language
- Japanese
- ISSN
- 0389-6633
2435-4295
The purpose of this study was to examine the effects of the built environment and social capital on health status over three years using dimensional compression and structural modeling of information on the living behavior of older people using machine learning. The study design included conducting a prospective cohort study using a probabilistic latent semantic structure model. As a result, we estimated that if the built environment and social capital were improved from the highest health risk conditions to the lowest health risk conditions, health risks such as the mortality would be suppressed to about 0.9 times compared to average health risks.